Abstract Autism spectrum disorder (ASD) is a set of neurodevelopmental disorders marked by impaired social communication and interaction, and repetitive, restricted behaviors and interests. Studies from the U.S. and Europe suggest that prenatal air pollution exposure (PNAPE), and exposure during early infancy, increase risk for ASD. PNAPE-associated risk for ASD has not yet been well studied in areas with very high levels of air pollution around the world. This proposal capitalizes on a once-in-a-life-time opportunity afforded by the quantifiable manipulation of air quality in Beijing, China during the 2008 Olympics and Paralympics. A series of aggressive pollution-control measures, implemented from July 20 to September 19, 2008, resulted in drastic reductions in air pollutant concentrations during this well-defined period. The marked temporary improvement in air quality within this 2-month time window provides a unique natural experiment, enabling us to explore population-wide effects of PNAPE on ASD during specific stages of prenatal (and early post-natal) development in a large urban population of children. This project will first characterize quantitative ASD traits and estimate ASD prevalence. Subsequently, the association of PNAPE with ASD will be evaluated, assessing ASD as both a categorical diagnosis and as a continuous trait. The proposed study will advance our scientific understanding of ASD trait distribution in a well-defined population, elucidate the effect of PNAPE on ASD risk, examine whether specific PNAPE windows differentially alter ASD risk, and determine whether birth outcomes mediate the associations. The specific aims are: 1) Estimate population-based prevalence of ASD in Beijing City among children who were in utero before, during, and after the Olympic period; 2) Assess time-window- specific PNAPE effects on ASD; and 3) Investigate the extent to which the effect of PNAPE on ASD and ASD traits is mediated by birthweight and gestational age at birth. We will also show the feasibility of establishing a biosample repository in this population by collecting saliva samples and accessing banked cord blood. This will enable future investigations of genetic or epigenetic mechanisms involved in PNAPE associations with ASD and related quantitative traits. The investigative team is uniquely poised to exploit the rare population-wide natural experiment of a massive air quality intervention. We have experience and expertise in ASD screening and assessment in Chinese populations, with surveillance and trait psychometrics, and in air pollution assessment. This study explores the potential for reducing early-life air pollution exposure as a public-health intervention to lessen ASD risk.